The Japanese government has put forth its "Space Industry Vision 2030" in hopes of expanding applicability beyond the space equipment industry to other relevant industries as well, with the aim of fast-paced doubling of the overall space industry's size by 2030. Amidst these circumstances, the Ministry of Economy, Trade and Industry launched Japan’s first satellite data platform “” intended for industry use on February 21, 2019. facilitates the use of traditionally difficult-to-handle satellite data by everyone from private companies, universities and research institutions through to individuals while also aiding in the creation of businesses in a broad range of industries and business categories made possible with this free access to space data.
The Tellus Satellite Challenge is a data analysis contest aimed at promoting the use of , namely in the form of producing a visual representation of satellite data use case models, uncovering outstanding human resources specializing in data analysis, and disseminating information and educating others on the types of satellite data and its various formats. This is the 4th time the event has been held.
The theme of this fourth event is the “Coastline extraction using SAR data.”
As an island nation surrounded by oceans, Japan holds a coast line which nearly has 35,000km in total extension. This is the 6th length in the world, and the 2nd after Philippines for the coast length per land area. Also, it is said that the total beach area in Japan is now 19,000ha, but coastal erosion continues on some coasts after the disappearance of about 13%( which is about 2,400ha) of the coast in the 15 years after 1978. If most of the beaches disappears, the reproduction will be a huge cost.
A coast plays an important role in storm surges and tsunami disaster prevention, by weakening the power of the wave. Also, it will affect the plants & animals that use the coast as a habitat. The serious coast erosion in all over the country occurs when the normal supply cycle of the sand collapse. It is caused due to disasters like earthquakes and typhoons, the change of directions of waves and wind due to the change of the weather, sea level rise due to global warming, and due to human factors like port improvement or dam constructions.
Now, the Japanese government is monitoring all of the coast in the nation by a satellite, to study to discover the signs of beach erosion as soon as possible, and prevent it before it gets too late. By using a satellite, it is possible to monitor a wide range of the coast area with high frequency. But the reality is, that a special skill is required to find out the coast area from the observed data. Therefore, in this challenge, we will like to develop an algorithm that can extract a coast line in a high accuracy from a SAR data.
We will use the observed data of PALSAR-2, loaded on the Advanced Land Observing Satellite “ALOS-2” which is for part of the coast in Japan. PALSAR-2 is a sensor called Synthetic aperture radar (SAR) that irradiates a radio wave to the ground surface and gains the information by the reflected radio wave from the ground surface. The feature of the SAR is that it can record the surface day or night, and also regardless of the weather. Check here for details.
Extract the coastline from the L-band SAR data.
・L-band SAR image
Satellite: ALOS-2 (JAXA)
Size: 600×600～10000×10000 (Depends on the image)
resolution: 3m × 3m
HH polarization, wavelength: About 24 cm
Number of coasts: 17
Number of images: 55 images, (Train) 25 images、(Test) 30 images
*Polarization is one indicator of the properties of radio waves and indicates the direction of vibration of an electric field. HH transmits horizontally polarized waves and receives horizontally polarized waves.
This article summarizes reference papers regarding coastline extraction.